\(\color{darkblue}{\textbf{Data Basics}}\)
\(\color{dodgerblue}{\textbf{Process}}\)
\(\color{dodgerblue}{\textbf{Analysis Hierarchy}}\)
\(\color{dodgerblue}{\textbf{Equity}}\)
- Where does this data come from?
- Why was this data collected?
- How was this data generated?
- Is this data demographically representative?
- Who is included and who is excluded from this data?
- Whose voices, lives, and experiences are missing?
- How much can this data be disaggregated by race, gender, ethnicity, etc.?
- Are the categories mutually exclusive and fully inclusive?
- Are there “other” categories and, if so, who does that include?
- Who stands to benefit from this data?
- Who might be harmed by the collection or publication of this data?
(See more in Urban Institute’s Do No Harm Guide)
\(\color{dodgerblue}{\textbf{Troubleshooting}}\)
| Issue | Common Fix |
|---|---|
| Input errors | Cleaning, omitting |
| Unrealistic observations | Sanity checks, filter thresholds |
| Noise/measurement errors | Averaging, interpolation, de-noising |
| Low/heterogenous density | Spatio-temporal aggregation, re-zoning |
| Representativity/biases | Normalization, acknowledgment, additional data collection |
\(\color{darkblue}{\textbf{Data Wrangling}}\)
\(\color{dodgerblue}{\textbf{Data Sources}}\)
\(\color{darkblue}{\textbf{Visualization}}\)
\(\color{dodgerblue}{\textbf{R}}\)
\(\color{dodgerblue}{\textbf{Python}}\)